The study developed deep‑learning pipelines using B‑mode and shear wave elastography (SWE) ultrasound images to stage fibrosis and identify at‑risk metabolic dysfunction‑associated steatohepatitis (MASH) in patients with MASLD. Across 250 examinations, SWE‑based learning consistently outperformed B‑mode learning, achieving higher AUROC scores for fibrosis stages F≥2, F≥3, and F4. End‑to‑end SWE models matched operator‑guided SWE performance for fibrosis staging.
arXiv:2607. 12062v1 Announce Type: cross Abstract: Differentiating non-alcoholic steatohepatitis (NASH) from non-alcoholic fatty liver disease (NAFLD) using ultrasound remains challenging due to subtle tissue alterations and the limited information available in conventional B-mode imaging.
The paper introduces an imaging-based method that uses large-scale computer vision models to analyze routine abdominal ultrasound images for predicting cirrhosis decompensation. It extracts predictive features beyond traditional laboratory risk scores, offering a non-invasive, low-cost, and scalable approach for early risk stratification. The framework combines automated ultrasound processing with modern deep learning to identify high-risk patients before clinical deterioration occurs.
DINO-Med introduces a patch‑based framework that adapts natural‑image foundation models, specifically DINOv3, to multi‑modal medical imaging. The method uses training‑free registration, automated localization, and mask‑filtered patch extraction to aggregate patch‑level features into subject‑level diagnostics. In liver fibrosis staging, DINOv3 outperforms handcrafted radiomics, ResNet, and SAM‑Med2D features, achieving 78.4% accuracy for mild fibrosis (S1) and 75.8% for cirrhosis (S4) on the CARE 2025 cohort.
The study evaluates lesion‑guided region‑of‑interest (ROI) deep learning for ovarian ultrasound classification, comparing it to global image, lesion contour, and contour‑based radiomics approaches across two public datasets. Using four deep‑learning architectures, the lesion‑guided ROI strategy achieved the highest accuracy (93.10% on MMOTU and 97.56% on OUD) with an AUC of 0.99, while requiring less annotation effort than contour‑based methods.
The study introduces dLI-RADS, an interpretable AI system that estimates hepatocellular carcinoma probability from multiphase CT images of liver lesions while aligning with LI-RADS imaging concepts. By combining neural network predictions with handcrafted quantitative imaging features and logistic regression, dLI-RADS achieved 80% lesion-level accuracy, outperforming standard LI-RADS (74%) in a multicenter retrospective cohort. The method was validated using bootstrap non‑inferiority tests and external sensitivity comparisons.
arXiv:2610.08528v1 Announce Type: cross Abstract: Clinical diagnosis is inherently a structured reasoning process, yet existing deep learning models often bypass this structure by mapping image featu...
The paper presents AppendiGrade, an XAI‑enhanced deep learning framework that automatically detects complicated appendicitis from ultrasound images. Using a dataset of 4,679 images across five classes, the authors trained four pretrained models and achieved a best accuracy of 95.58% with InceptionV3 after applying preprocessing, hyperparameter tuning, and image sharpening. Grad‑CAM heatmaps were generated to explain the model’s predictions, facilitating easier expert cross‑checking.
Background: Microvascular invasion (MVI) predicts recurrence and survival in hepatocellular carcinoma (HCC) but requires postoperative histopathology for diagnosis. We developed and validated a model...
The paper introduces DINO-Med, a patch‑based framework that adapts natural‑image foundation models to multi‑modal medical imaging, specifically for liver fibrosis staging. It processes raw multimodal scans through training‑free registration, automated localization, and mask‑filtered patch extraction, then aggregates patch‑level insights into subject‑level diagnostics. Using the CARE 2025 Liver Track 4 cohort, the DINOv3‑based approach achieved the highest classification accuracy (78.4% for mild fibrosis and 75.8% for cirrhosis) compared to other feature representations.
arXiv:2607. 08219v1 Announce Type: cross Abstract: The privacy requirements of medical data and its substantial variations across organs and modalities hinder the clinical implementation of medical AI.
arXiv:2607. 08219v2 Announce Type: replace-cross Abstract: The privacy requirements of medical data and its substantial variations across organs and modalities hinder the clinical implementation of medical AI.
The study developed a multimodal ultrasound and clinical data model to predict microvascular invasion (MVI) preoperatively in hepatocellular carcinoma (HCC). Using data from 489 patients across eight centers, the model combined B-mode ultrasound, color Doppler flow imaging, dynamic contrast-enhanced ultrasound, and clinical information, achieving an AUC of 0.8953 in external validation. Dynamic contrast-enhanced ultrasound contributed the most predictive power, while other modalities and clinical data added complementary value.